<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hello,<br>
<br>
I'm trying to parallelize a process using pvbatch and MPI, with
MultiBlock data set; thus using the vtk composite pipeline.<br>
I made a sample python program that is representative of what I have
to do :<br>
<br>
--------------------------------------------------------------------------------------------------<br>
<blockquote><i>from paraview.simple import *</i><i><br>
</i><i><br>
</i><i>r = servermanager.sources.XMLMultiBlockDataReader()</i><i><br>
</i><i>r.FileName = "input.vtm"</i><i><br>
</i><i><br>
</i><i># Defining a sample fake data processing<br>
nbTs = 1000</i><i><br>
</i><i>ts = {}</i><i><br>
</i><i>for tIndex in range( 0, nbTs ):</i><i><br>
</i><i> ts[tIndex] = servermanager.filters.Transform()</i><i><br>
</i><i> if tIndex == 0:</i><i><br>
</i><i> ts[tIndex].Input = r</i><i><br>
</i><i> else:</i><i><br>
</i><i> ts[tIndex].Input = ts[tIndex - 1]</i><i><br>
</i><i> ts[tIndex].Transform.Scale = [1.01,1.01,1.01]</i><i><br>
</i><i><br>
</i><i>w = servermanager.writers.XMLMultiBlockDataWriter()</i><i><br>
</i><i>w.Input = ts[nbTs - 1]</i><i><br>
</i><i>w.FileName = "output.vtm"</i><i><br>
</i><i><br>
</i><i>w.UpdatePipeline()</i><i><br>
</i></blockquote>
--------------------------------------------------------------------------------------------------<br>
<br>
I launch that using <i>"mpiexec -np 4 pvbatch myscript.py</i>"<br>
All run well but I get a longer time using MPI than using only "<i>pvbatch
myscript.py".<br>
<br>
</i>By monitoring RAM, I noticed that it seems the data is loaded on
time by MPI process, and (maybe) all the MPI processes do exactly
the same job, computing four times all the data.<br>
<br>
Why my blocks in MultiBlock data set aren't dispatched over the MPI
processes ?<br>
What am I doing wrong ?<br>
<br>
Many thanks for any help,<br>
<br>
Yves<br>
</body>
</html>